Web Scraping - Mission to Mars

---

A Responsive Web Visualization Dashboard using Flask framework to create a application that scrapes various websites for data related to the Mars and displays the information in a single HTML page.

📝 Table of Contents

• About
• Getting Started
• Hosting
• Acknowledgments

🧐 About

In building this dashboard based on two parts. First part consist of scraping the data from using jupyter noebook and designing a dashboard using templated in Flask Framework.

🏁 Getting Started

Step 1 - Scraping

Completed initial scraping using Jupyter Notebook, BeautifulSoup, Pandas, and Requests/Splinter.

• Created a Jupyter Notebook file called mission_to_mars.ipynb and used this to complete all of the scraping and analysis tasks.

NASA Mars News

• Scraped the NASA Mars News Site and collected the latest News Title and Paragraph Text. Assigned the text to variables that can be referenced later.

# Code:
news_title = "NASA's Next Mars Mission to Investigate Interior of Red Planet"

news_p = "Preparation of NASA's next spacecraft to Mars, InSight, has ramped up this summer, on course for launch next May from Vandenberg Air Force Base in central California -- the first interplanetary launch in history from America's West Coast."

JPL Mars Space Images - Featured Image

• Visited the url for JPL Featured Space Image here.

• Used splinter to navigate the site and find the image url for the current Featured Mars Image and assigned the url string to a variable called featured_image_url.

• Made sure to find the image url to the full size .jpg image.

• Made sure to save a complete url string for this image.

# Example:
featured_image_url = 'https://www.jpl.nasa.gov/spaceimages/images/largesize/PIA16225_hires.jpg'

Mars Facts

• Visited the Mars Facts webpage here and used Pandas to scrape the table containing facts about the planet including Diameter, Mass, etc.

• Used Pandas to convert the data to a HTML table string.

Mars Hemispheres

• Visited the USGS Astrogeology site here to obtain high resolution images for each of Mar's hemispheres.

• Used splinter to click each of the links to the hemispheres in order to find the image url to the full resolution image.

• Saved both the image url string for the full resolution hemisphere image, and the Hemisphere title containing the hemisphere name. Used a Python dictionary to store the data using the keys img_url and title.

• Appended the dictionary with the image url string and the hemisphere title to a list. This list will contain one dictionary for each hemisphere.

# Example:
hemisphere_image_urls = [
    {"title": "Valles Marineris Hemisphere", "img_url": "..."},
    {"title": "Cerberus Hemisphere", "img_url": "..."},
    {"title": "Schiaparelli Hemisphere", "img_url": "..."},
    {"title": "Syrtis Major Hemisphere", "img_url": "..."},
]

---

Tasks

• Used Splinter to navigate the sites when needed and BeautifulSoup to help find and parse out the necessary data.

• Used Pymongo for CRUD applications for your database.

• Used Bootstrap to structure my HTML template.

✍️ Hosting

Step 2 - MongoDB and Flask Application

Used MongoDB with Flask templating to create a new HTML page that displays all of the information that was scraped from the URLs above.

• Started by converting my Jupyter notebook into a Python script called scrape_mars.py with a function called scrape that will execute all of my scraping code from above and returned one Python dictionary containing all of the scraped data.

• Next, created a route called /scrape that will import your scrape_mars.py script and called my scrape function.

  • Stored the return value in Mongo as a Python dictionary.

• Created a root route / that will query your Mongo database and passed the mars data into an HTML template to display the data.

• Created a template HTML file called index.html that will take the mars data dictionary and displayed all of the data in the appropriate HTML elements.

---

🎉 Acknowledgements

• UWA Data Science